Swinburne's Tritheism
Swinburne's Tritheism
Swinburne's Tritheism
175
c 1997 Kluwer Academic Publishers. Printed in the Netherlands.
Swinburne’s Tritheism
EDWARD C. FESER
University of California, Santa Barbara, USA
So there would be no moral requirement that the three (perfectly good) divine
individuals would have to meet by bringing about a fourth divine individual,
and so no overriding reason for them to bring one about. And if there is no
overriding reason to bring about a fourth divine individual, there could not
be a fourth divine individual. For even if the three existing divine individuals
decided to bring about a fourth individual in some respects like them, despite
the fact that they would not be morally required to do so, it wouldn’t (and
couldn’t) be a metaphysically necessary individual, because it would not
have come about inevitably in virtue of the properties of the three divine
individuals. But then it just wouldn’t (and couldn’t) be a divine individual:
Such an individual ‘might (metaphysically) not have existed, and so could
not be divine’.12
With this argument, Swinburne claims to have shown a priori that ‘neces-
sarily if there is at least one divine individual, and if it is logically possible
that there be more than one divine individual, then there are three and only
three divine individuals’.13 But he has not, I think, quite shown this at all. First
of all, what he can, at best, claim to have shown is, not that if there are any
divine individuals at all, there are three and only three, but only that if there is
.just one divine individual at first, in some non-temporal sense of ‘at first’ that
fits in with Swinburne’s non-temporal sense of ‘bringing about’ (if we grant
the meaningfulness of this sort of talk at all), then that divine individual will
inevitably bring about two, and only two, further such individuals. But this
does not at all rule out the possibility that there just are, in the first place, any
number of divine individuals, who exist of metaphysical necessity. It could,
for all Swinburne has said, turn out that there ‘already’ are three, or three
thousand, divine individuals who exist without being actively brought about
by any other such beings, but who permit each other to exist inevitably in
virtue of their properties (i.e., because it is morally required that they permit
this). (And in this case, presumably, no further divine individuals, a three
thousand and first, say, could actively be brought about because there would
be no overriding need to do so. For there would already be, for each divine
individual, others to ‘share in love’ with.)14
Now Swinburne might be able to grant this point without undermining
his overall project. He could acknowledge that he hasn’t, after all, shown
that there couldn’t be more than three divine individuals, but only that, if we
are willing to grant the existence of one such individual, we must grant the
existence of two more as well. And he would then presumably argue that the
inductive arguments which he elsewhere argues make it probable that there
is a God15 warrant only the conclusion that there is one such being.16 So his
overall argument could be reformulated as follows: Inductive considerations
by themselves warrant only the conclusion that there is one divine individual;
SWINBURNE’S TRITHEISM 179
but a priori considerations show that, if there is at least one divine individual,
there are three of them; so our overall evidence warrants only the conclusion
that there are three divine individuals.17
Still, Swinburne’s argument has, in my view, other, more serious problems
than this one. For it rests on what can only be described as rather tendentious
moral intuitions. Why should we accept Swinburne’s claim that the states
of affairs involving one individual sharing with another, and two individuals
sharing with a third, but not the state of affairs where three individuals share
with a fourth, are states of affairs a divine individual has a moral obligation
to bring about? Swinburne appeals to the analogies of marriage and family to
argue for the moral necessity of bringing about the first two states of affairs,
but it is easy to see how they would also support the moral necessity of
bringing about the third. One could argue as follows: ‘We know from human
experience that a child who has no siblings often feels that his family life is
less fulfilling than that of children who do have siblings; so a state of affairs
in which love is shared between four individuals is better than that in which
it is shared between three. So a perfectly good divine individual, since he is
obligated to bring about the best state of affairs possible, must bring about
three other divine individuals. In fact, since very large families (including
as many as ten children) are often the happiest, perhaps we should conclude
that there must be, after all, eleven or twelve divine beings’. I do not claim
that such an argument would be very convincing. But it is not obvious that
it would be any weaker than Swinburne’s argument. Swinburne’s intuitions
are, I think, far too dubious to support the massive metaphysical weight he
wants to rest on them.18
But, even apart from these worries, we might also ask: Is it logically possible
that there exists more than one divine individual in the first place? Can it even
coherently be supposed that there is? One reason for thinking not is that
it is arguable that there could not be more than one omnipotent individual,
for if there were two or more allegedly omnipotent individuals, they could
conceivably frustrate one another’s actions, in which case they just wouldn’t
truly be omnipotent at all (since an agent whose actions can be frustrated
is by definition not omnipotent). Swinburne considers this objection, and
thinks that he can get around it.19 He suggests first that, since the two or more
omnipotent divine beings would also be perfectly good, there would be a large
range of actions concerning which they could not possibly conflict. Each will
recognize a certain range of actions as morally required, and a certain other
range of actions as morally prohibited, and thus will agree on the need to
perform or refrain from such actions. But what about acts that are neither
morally required or morally prohibited, but which a divine individual may
choose to do or not do as he wishes? What if one divine individual decides
180 EDWARD C. FESER
on his construal, the logical consistency of (l)–(7) is obvious. What, then, has
all the fuss been about? Have Davis, Martinich, van Inwagen, Cartwright,
Macnamara et al. (all cited in note 2) and many others all been committing an
embarassingly colossal blunder in even thinking that there was any question
as to whether Trinitarianism is coherent? That is hard to believe. Yet they
must have been if Swinburne is right. Of course, it could be that he is right,
and that all those authors (many of whom are themselves Trinitarians) who
have supposed that the correct reading of the creeds raises questions about
the coherence of Trinitarianism are just mistaken. But this does not seem
likely, and to convince us of his position, Swinburne would surely have to
present detailed historical, philological, and theological arguments that go
well beyond the few remarks he makes in defense of it in The Christian God.
As to Swinburne’s claim that the only other possible reading of the creeds
entails a contradiction, I presume that most of the writers just mentioned
would disagree with it. Whether it is true or not, I leave to the reader to
decide.
We see, then, that Swinburne’s account fails, not only as an attempt to
demonstrate the coherence of the doctrine of the Trinity, but even as an
attempt to state the content of the doctrine. For if Swinburne’s a priori argu-
ments concerning the nature of God were convincing, they would, it turns
out, constitute a strong case for Tritheism, and thereby serve to undermine
Trinitarianism. This is, no doubt, not a result Swinburne would be comfort-
able with. It is, then, perhaps fortunate for him that those arguments are not
convincing.27
Notes
says, there simply could not be a (logically possible) action an omnipotent individual is
frustrated in performing. So the suggestion that two omnipotent individuals might frustrate
one another’s actions is incoherent.
It seems to me, though, that the two cases are not quite analogous. The paradox of the stone
is intended to be a reductio ad absurdum of the notion of omnipotence, and the argument
can be stated in the form of a dilemma: ‘Suppose there is an omnipotent being. Can he
create a stone that is too heavy for him to lift? If he can, there is something he cannot do,
namely lift the stone in question, so that he is not omnipotent. If he can’t, then there is
something he cannot do, namely create the stone, so that he is not omnipotent. Either way,
he is not omnipotent. So there could not be an omnipotent being’. The problem with this
argument, as we have seen, is that it depends on the idea that omnipotence involves the
ability to do the logically impossible. But there is no analogous problem with a reductio
argument against the possibility of more than one omnipotent individual, which can be
stated as follows: ‘Suppose there are two omnipotent individuals A and B. Now (using
Swinburne’s example, The Christian God, p. 172) suppose A desires to send Abraham to
Iraq at time t and that B desires to send Abraham to Iran at time t. Since A is omnipotent,
he can bring it about that Abraham is in Iraq at t; and since B is omnipotent, he can bring
it about that Abraham is in Iran at t. It follows, then, that Abraham can be in both Iraq and
Iran at time t. But this is impossible. So there cannot be two omnipotent individuals’. This
argument, unlike that concerning the paradox of the stone, does not depend on the idea
that an omnipotent being can do the logically impossible. So, contrary to Anderson, there
seems no reason to doubt that it constitutes a genuine difficulty for Swinburne’s position,
one which, as we have seen, he fails adequately to deal with.
21. Swinburne, The Christian God, p. 180.
22. Swinburne, The Christian God, p. 180.
23. Swinburne, The Christian God, p. 180.
24. Swinburne, The Christian God, pp. 181–182.
25. Quoted, along with much of the rest of the creed, by Swinburne in The Christian God,
p. 186.
26. Swinburne, The Christian God, p. 181.
27. I thank C. Anthony Anderson, J. William Forgie, and an anonymous referee for comments
on earlier versions of this paper.